HIV-1 associated encephalopahty which may develop into severe cognitive impariments refered to as the AIDS dementia complex causes damage in the human brain through both direct and indirect mechanisms. An example of HIV-1 non-structural proteins which participates in brain pathogenesis is the viral tat protein. The ability of HIV-1 proteins, particularly tat, which is a transactivating protein to regulate cellular functions helps explain the dysfunction of the nervous system in brain tissue where there is little evidence of active virus multiplication. We found that HIV-1 up regulates the synthesis and release of the beta-chemokine MCP-1 which is also found in elevated levels in the CSF of AIDS patients with dementia. There was also elevated levels of mRNA to MCP-1 in brain tissue from demented patients. Investigations of CSF samples from HIV-1 infected non-demented patients as well as non neurologically impaired patients did not show increases in MCP-1 in the CSF. This observation has now been confirmed by several other laboratories and may serve as a surrogate marker for AIDS associated dementia. We have also shown that human astrocytes are responsible for MCP-1 release in a model of the blood:brain: barrier with endothelial cells. Also the MCP-1 released chemoattracts monocytes across the barrier and upregulates the beta-chemokine HIV-1 co-receptor, CCR5, on migrating monocytes. The molecular regulation of MCP-1 in astrocytes may be controlled by transcription factors which also regulate JCV infeciton, namely NF-1. The promoter sequences of the human MCP-1 promoter shows inducible NF-1/AP-1 sites which are sensitive to the HIV-1 protein tat. We are continuing to assay a number of wild type and deletion mutant MCP-1 promter expression vectors to identify the critical DNA binding regions which are responsive to tat up-regulation in human astrocytes. It appears that sequences from -213 which contain the NF-kB and Sp1 protein binding sites and sequences -300 to -420 which contain NF-1 binding sites are active in MCP-1 mRNA synthesis upon tat stimulation. We had previoulsy established a cell culture model of HIV-1 infection in human astrocytes closely resembling a viral latency. Upon treatment of HIV-1 latenly infected cells with proinflammatory cytokines like TNF-a, the viral genome is activated and new progeny virions are released. We have now shown that IL-1 beta and the tat also activate new virus synthesis. This infectious process however is not cytopathic to the cells which suggests that the astrocyte may serve as a reservoir for HIV-1 in the brain. We have also shown that clinical isolates of HIV-1 from patient CSF or blood are able to infect astrocytes in an CD4 and/or chemokine co-receptor independent manner. We also have determined that sequences of the viral gp120 glycoprotein in the V3 loop are not involved in binding to astrocytes unlike binding to lymphoid cells. With the emphasis on the development of an AIDS vaccine, understanding the mechanims of HIV-1 infection in human brain is critical to the performance and either prophylactic or therapeutic use of a vaccine. If reactivation of a latent infection from a sequestered site such as the brain can take place, immune reactivity as a result of vaccination would be necessary to clear the neuroinvasive virus as well as virus in the periphery. Considering the diverse nature of HIV-1 strains and their re-occurrence, it is necessary to investigate both the neruovirulence of HIV-1 and the emergence of neurotropic strains.